Experimental Setup: 2M226 CW Magnetron
Document Experiment
Over the last several years I decided to do some organizing of my electronic components; as a result several old microwave ovens
were disassembled for parts; winding up with several oven magnetrons to experiment with. Operating these CW magnetrons in the
CW mode is in my opinion extremely dangerious unless you really know what you are doing; afterall they are designed for cooking! Operating them
in the short duratrion pulse mode was the approach used. This page briefly describes the experimental setup and results. Some detail
is given for the magnetron unit is given on thes page; The 2M226 pulse modulator component assembly details is given on another
page. Construction details are given the in following PDFs: in future links
The two photos below shows the magnetron head assembly RF exciting an Neon bulb placed in front of the waveguide.
Components Identified/ Entire System
2M226:
Frequency => 2460 Mhz
Peak Anode Voltage => 4.2 KV
Anode Current => ~ 300 ma.
Heater 3.3V at 11 A
As can be seen from the above photos the magnetron was not properly loaded at its output (open waveguide) I didn't have a proper load
available; which could explain why during its operation the cathode voltage appeared to be very unstable ( mode switching) especially
near its operating voltage.
I didn't have any current sensors available at the time to monitor the the anode current; howeveras can be seen from the photo (right) the peak cathode voltage was monitored. The peak pulse was not flat at the magnetron's
intended 14Kohm operating impedance. Its been a while since I tested this unit so I don't recall how flat the modulator
tested with a 14Kohm resistive load at the the secondary of the step up transformer; generally it is if the pulse transformer has
adequate volt - sec. At this point in time I suspect the transformer not having adequate core.
The initial portion of the pulse
edge is rather slow because the pulse modulator uses an SCR array for the PFN HV switch; which is relatively slow of several hundred
nanoseconds for the low cost SCRs. Also there may be too much inductance slowing things down. I plan to check this out in the future.
The RF envelope was monitored by connecting the DSO probe leads across a 1N23 microwave diode (right). Near the peak, the pulse
envelope shape resembles the above cathode peak. Assuming that the leading edge peak is round due to the slow leading cathode
edge; then reaches its peak and falls down because the cathode voltage falls.
Magnetron Assembly
(see below more details)
2M226 Pulse Modulator/ PFN Assembly
(future link)
14 VDC Magnetron Heater Power Source
Pulse Modulator Power Source
magnetron heater power input
magnetron heater switch mode power supply
cooler for Shottky full wave diode for filiament PWR
cooler for IGBT filiament transformer drive
filiament transformer ; Litz wire for secondary